nvlabs / fermat

Fermat is a high performance research oriented physically based rendering system, trying to produce beautiful pictures following the mathematician’s principle of least time

#Fermat

Fermat 2.0 is a CUDA physically based research renderer designed and developed by Jacopo Pantaleoni at NVIDIA. Its purpose is mostly educational: it is primarily intended to teach how to write rendering algorithms, ranging from simple forward path tracing, to bidirectional path tracing, to Metropolis light transport with many of its variants, and do so on massively parallel hardware.

The choice of CUDA C++ has been made for various reasons: the first was to allow the highest level of expression and flexibility in terms of programmability (for example, with template based meta-programming); the second was perhaps historical: when Fermat's development was started, other ray tracing platforms like Microsoft DXR did not yet exist. The ray tracing core employed by Fermat is OptiX 6.0 - though future versions might also add a DXR or VKRT backend.

Fermat is built on top of another library co-developed for this project: CUGAR - CUDA Graphics AcceleratoR. This is a template library of low-level graphics tools, including algorithms for BVH, Kd-tree and octree construction, sphericals harmonics, sampling, and so on and so on. While packaged together, CUGAR can be thought of a separate educational project by itself. More information can be found in the relevant Doxygen documentation.

The official project page can be found here: https://nvlabs.github.io/fermat/

DEPENDENCIES:

Fermat has the following dependencies:

• cub : contained in the package
• freeglut : contained in the package
• Assimp : contained in the package
• CUDA 10.0 : not contained - it should be separately downloaded and installed on the system
• OptiX 6.0 : not contained - it should be separately downloaded and copied in the folder contrib/OptiX

Its distribution also contains a set of adapted models originally taken from Benedikt Bitterli's rendering resources: https://benedikt-bitterli.me/resources.

COMPILATION:

Once all dependencies are sorted out, the Visual Studio 2015 solution file vs/fermat/fermat.sln can be opened and the project can be compiled.

USE:

After compilation, you can launch Fermat's path tracer with the following command line: fermat.exe -view -pt -r 1600 900 -i ../../models/bathroom2/bathroom.fa

fermat's Issues

Can't get the view

Got it working with WIN10 October Update, Optix 5.1.0 and CUDA 10 after a good old manual VS hacking :D But I can't get the live window, only the results on the terminal/console ->

https://i.postimg.cc/kXbd5Y72/Untitled.jpg

Where's a damn score? (average FPS/average Rays per second, SOMETHING please! ) Ah, the bathroom file is zipped (there's a ZIP, not an OBJ in the dir :p ) ......please correct (and maybe update to CUDA 10 + Optix 5.1.0) the package on github ;)

Support CUDA 10.1 SDK..

Hi,
fermat has issues building with CUDA 10.1 SDK..
can post detailed build failures but at least is failing with missing thrust functions..
a fix will be nice..
thanks..

Support shader models?

I can't compile Fermat 2.0 with sm_61,compute_61 for GTX 1070:

3>D:\Projekte\3D\NVidia\fermat\src\pathtracer_core.h(550): error : identifier "__match_all_sync" is undefined

Is this expected and Fermat should only work with RTX cards?

ptxas fatal compile error

1>ptxas C : /Users/phil/AppData/Local/Temp/tmpxft_00002220_00000000-6_optix_shadow_material.ptx, line 522; error : Label expected for argument 0 of instruction 'call'
1>ptxas C : /Users/phil/AppData/Local/Temp/tmpxft_00002220_00000000-6_optix_shadow_material.ptx, line 522; error : Function '_rt_ignore_intersection' not declared in this scope
1> ptxas C:/Users/phil/AppData/Local/Temp/tmpxft_00002220_00000000-6_optix_shadow_material.ptx, line 522; fatal : Call target not recognized
1> ptxas fatal : Ptx assembly aborted due to errors
1>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V140\BuildCustomizations\CUDA 10.0.targets(712,9): error MSB3721: The command ""C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\bin\nvcc.exe" -gencode=arch=compute_52,code="sm_52,compute_52" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\bin\x86_amd64" -x cu -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\include" -I../../src/ -I../../src/mesh/ -I../../contrib -I../../contrib/OptiX/include/ -I../../contrib/freeglut/include/ -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\include" -I../../src/ -I../../src/mesh/ -I../../contrib -I../../contrib/OptiX/include/ -I../../contrib/freeglut/include/ -I../../contrib/cub -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\include" -G --keep-dir x64\Debug -maxrregcount=0 --machine 64 --compile -cudart static -g -D_DEBUG -DWIN32 -D_CONSOLE -D_CRT_SECURE_NO_WARNINGS -DNOMINMAX -D_USE_MATH_DEFINES -D_UNICODE -DUNICODE -Xcompiler "/EHsc /W3 /nologo /Od /Fdx64\Debug\vc140.pdb /FS /Zi /RTC1 /MDd " -o x64\Debug\optix_shadow_material.cu.obj "C:\Users\phil\Downloads\fermat-master\src\kernels\optix_shadow_material.cu"" exited with code 255.

This error is happening on optix_shadow_material.cu, optix_rt.cu, and optix_material.cu.

What am I doing wrong?

Project (from description) seems Interesting.. some ETA of source code drop?

Document settings used to generate images in docs

Any chance we could get the settings, number of progressions etc. used to generate the images in the docs? That way we can verify that our build is working correctly and haven't broken anything. For example I'm finding it difficult to reproduce the water caustics image on the BPTLib page.

Missing files?

It appears there's many files missing.

For example:
contrib/cugar/basic/numbers.h
includes
#include <vector_types.h>
#include <vector_functions.h>

These files don't exist in the repository.

Linux support

Hi, I'd like to get fermat working on Linux, would you be interested in supporting build via Makefile? Is there any show stopper that would preclude that from being possible?